July Mailbag

Time for a big Q&A!

Thanks so much for all the questions posted on Twitter - to be honest I was a little bit overwhelmed with how many I got and how interesting many of them were to discuss! Apologies if I didn’t get to your question, but I really appreciate every single person taking the time out to ask.

* Thanks so much to everyone who has bought me a coffee. It is much appreciated! *

Bazball Convert: “As a cricketing data analyst, what has been the biggest lesson you have learnt from your work with Punjab, Leicester and Birmingham Phoenix?”

Love this question. Firstly, working with different personalities and how to ‘manage upwards’, if that makes sense. I’ve been pretty lucky in the IPL/Hundred that the management groups were very open to evidence-based, data-driven points of view and as long as you could do that, you’d have a voice. Nico is great as well at Leicestershire.

Next up - affirmation of your methodology. When you first go into a team, and you’ve never played cricket before at any level and you're trying to influence decision-making of a person who has been involved in the industry for decades, it’s easy to fall into the trap of ‘imposter syndrome’. However, when you earn a coach/player’s trust and they come to you for your insights, and they value the work that you can provide, then you realise that you’re definitely not an imposter!

On a related subject - I also learnt that most players/ex-players aren’t geniuses, although there are some who are very, very astute, and most players shouldn’t be anywhere near recruitment - 99% of the time they overvalue the players who have played well against them. Same for most ex-players too, they’ll often have subjective bias based on what they’ve seen on TV or what someone else tells them.

The final learning, which hopefully will be useful for budding analysts looking for roles, is that taking the first job is vital. That then allows you to start building up a network of contacts who can help you advance your career. For example, if you work well with a coach, then there’s a good chance you go with them again in their next job. I can remember writing hundreds of emails and Linkedin messages to people in the industry when I first started, barely getting any replies. Actually the time would have been better spent offering to work for free (and I accept I was lucky enough to be in the position where I could have done that) and building up a contact list that way.

Ollie: “How can England produce more quality t20 bowlers?”

There’s no doubt that this is a major issue which England and coach Matthew Mott need to address in the short-term and longer-term future - England have a player pool which is extremely batting strong, but bowling weak, and you can see evidence of that with a number of picks in this year’s Hundred squads too.

I actually don’t think that there is a lack of quality T20 spinners in England. Players such as Callum Parkinson, Jake Lintott and Rehan Ahmed have evident quality and particularly in the case of the first two players, enough statistical evidence to suggest that they should be a very good franchise bowler, and indeed, with the potential to step up to international cricket. Rehan, having played one season so far, doesn’t have a huge sample of data but having worked with him during the summer, I can say for sure that he is a player of gigantic potential, as well as already being a high quality T20 spinner.

Pacers are more of a problem. A few have done reasonably well in overseas T20 leagues but the fact that England don’t have pacers who are considered strong options by IPL teams is a big issue - because if they were good enough, they would be. What I think can help is greater exposure in higher quality and higher profile leagues than the Blast - The Hundred will help with this but also the increasing number of short-format leagues worldwide will help as well. For example, there appears to be a big clash this winter between the new Emirates League, Big Bash, BPL and the new South Africa League, so what that will do is increase the pool of players playing franchise cricket this winter, and give opportunity to players who really need that greater exposure.

Jonas: “given the same amount of sample size of balls faced, which of these stats are more predictive and in what order -- balls/6, balls/4, non-boundary SR, dismissal rate.”

For me, dismissal rate is the most noisy here - for example let’s say there’s a new batter on the T20 circuit who has faced 200 balls and has been dismissed 10 times. They have a dismissal rate of 20 balls per wicket, which is around average. But what we don’t know is the context of those balls - have they been dropped five times, or played a ton of false shots? So that, in my view, has the potential to be more volatile in smaller samples.

Due to this, boundary percentage is much more usable in this case of that player and helps you to profile them, at the very least. If, from 200 balls, they have a boundary percentage (4+6) of 12% and a six percentage of 2%, you can be pretty certain that they aren’t a good hitter and lack power and intent - even in those 200 balls there are pretty no instances where this player will be a disguised above-average hitter, they just can’t be.

Using a more extreme example, let’s say there’s a 20 year old batter playing in their first T20 tournament and they’ve faced 80 balls, hitting 8 sixes (10% of balls faced). One thing you can guarantee at least from this sample is that they’ve got power and intent, and from that, at least some future upside, and at best, a potential superstar.

The Sceptical Cricketer: “Who are your top 4 English spinners for the World Cup? (Can include Howell given his middle over dominance)”

Rashid, Howell, Lintott, C Parkinson

James McCaghrey: “Do you think Englands aggressive batting mindset is sustainable in tests?”

I’m unconvinced. At this point in time, circumstances have conspired to favour England - New Zealand came into the series without much preparation and not quite at full strength, while it was tough for India coming back to play one Test too.

The concept of a ‘new manager bounce’ is well discussed in football, and this could well be the case here too. I don’t think there’s a high chance that a Silverwood/Root management group would have achieved close to these results.

We also have to consider that conditions in England have been unusually batting-friendly this summer, and so there’s still plenty of doubt in my mind as to whether England’s approach is sustainable in trickier conditions - such as away in Australia or in the subcontinent.

Broken Cricket Dreams Cricket Blog: “Is South Africa's ODI series withdrawal an indication for things to come?”

Yes, I think so. Bilateral series simply aren’t financially sustainable for many countries, and they can have the potential to make more money hosting, for example, a major T20 franchise league, instead. Traditionalists, of course, will hate this, but what do they think will happen to longer format cricket if a board goes bankrupt?

On a fairly related subject as well, I’m in full agreement with R Ashwin’s recent comments on 50 over cricket - simply from a fan’s perspective I find the format much more boring than a T20 match.

Biased Cricketing Opinions: “Do team analysts collect data for batter’s vs high pace? Like 140+ category v a medium pacer category.”

Yes, but the match has to be televised to have that data. But even without having full data, this can be useful to work out batter’s strengths and weakness.

Thomas Laver: “Who's the least well known player who's most likely to have a breakout Hundred season this year? Men's and women's sides!”

I haven’t watched much women’s domestic cricket this year, but I was super impressed with Alice Capsey last year in The Hundred, and I know Marie Kelly from Phoenix last year and she’s got a lot of runs at very high strike rates this summer also.

In the men’s, there’s a few I think can have impressive seasons. For my team Birmingham Phoenix, the young trio of batters - Smeed, Mousley, Benjamin - have big upside, and I also think Rehan Ahmed can make a name for himself on the big stage.

Mask: “What do you think of The 6ixty and such innovations? Are there any rule changes you're excited to see and/or adaptable into T20?”

I’m all for innovation and new formats. I’ve worked in The Hundred and T10, and working on new formats present many challenges, which I found both fun plus intellectually stimulating (e.g. working on par scores for a new format).

From reports, the BBL are going to reduce the rule changes for this season which I think is a shame. Teams were really slow to catch on to the value of the bash boost point, for example.

I like stuff like a Powerplay point - think this would be good to watch and help with intent in the first six overs at a time where in T20, teams are often trying to value wicket preservation - and I also think a substitute from the match day squad (so, for example, a choice of the 3 players remaining from a 14 player squad) which can be used at any stage of a match. If a team hits 220 batting first, they can bring on another bowler, for example - or if a team needs 40 from the last 3 overs, they can sub on a hitter for a bowler. Or, a team might simply value having a specialist fielder on the bench.

With teams often pushing the line for replacing a starting XI player in the field with a better fielder, and the advent of tactical retirements, having an any-time substitute reduces the need for those two aspects of the game which some people don’t consider in line with the spirit of cricket.

Stuart Akister: “Should Englands 20/20 and ODI teams vary much more than they do? We always seem to have a similar line up for both teams”

Absolutely. There is a continuing divergence between T20 and other formats, and I think England (and other teams) often make a big selection mistakes by seemingly just considering it a white-ball setup.

Michael Ash: “Are we still waiting for a tactical revolution in ODIs (similar to the level of tactics we get in T20), and what will it look like when it arrives?”

Fairly related question, and yes, I think so. ODI teams still have been surprisingly reticent to follow England’s lead by playing an ultra-aggressive brand of cricket, and moving towards this, and developing players capable of such an approach would be a start. I’m also surprised how few death-over specialists are used in ODIs.

Sarah: “Can we [England] win on Sunday?”

I think anything is possible with both of these teams - we’ve seen that already this summer. Both teams are going to need short and medium-term transitions in terms of selection, and I think there are legacy players on both sides whose time playing at the highest level could well be finished. However, both teams appear to (at least to some extent) be picking based on reputation as much as current ability.

England not having a current national selector doesn’t help either. Reports suggest this role will be re-created, but I’ve not yet seen any advertisements for it.

Shahir Rahman: “Hey Dan one from before, but who should Bangladesh pick for t20s? Current strategy is not working.”

I’ve written about Bangladesh before, but I don’t think much has changed. In my view, they need to overhaul their talent production, with the goal of starting to produce much more high intent power hitters, which I think they lack. It’s going to be a long-term process, but that’s what they should move towards. The pitches there often don’t help too - it might get them home victories but they aren’t similar to conditions they’ll likely face in major tournaments.

Soham: “What is your stance on the fast- tracking of tearaway pace bowlers like Umran Malik to their National teams ? Having seen such talents being dropped eventually for getting smashed and never coming back from it has really made me question this cycle of events.”

I agree. I’d much rather see a greater sample size of evidence that a player is good enough for the step up. In the case of Umran Malik, for example, I’d like to see him bowl much more outside of the middle overs before playing international cricket. I’m very surprised that he seems ahead of Arshdeep Singh (who regularly bowls 2 Powerplay/2 Death overs) in the India selection picking order.

Seb: “Hi Dan, do you use different metrics for different tournaments or conditions or do they stay pretty consistent regardless? For example would the way you identify players for the Hundred differently than you would for the IPL?”

Most of the time pretty similar but I use a model which values the quality/difficulty of batting/bowling in each tournament which varies according to each event. Also, the obvious difference in the Hundred compared to T20 leagues is there are 20 fewer balls, so you have to look at slightly different things - you might be even more unwilling to look at players capable of playing match-losing innings, because even 25 balls is a quarter of your batting resources.

Also, another Hundred example might be batters who are strong against all bowling types might be worth more, because if a player has an obvious weak match-up, the double over in the Hundred means that they can be exposed much more frequently than they could be in T20.

Bharathr: “How do cricketers react to data ?”

Depends on who it is! Younger players tend to be more open to it than veterans, although I’ve had some great conversations with players in their mid 30s as well. At the moment, I only tend to really go into deep data-driven discussions with players who ask for it, but as is the case in football, I can see a scenario developing where cricketers get it forced on them by management groups too, because they need to be accountable for performance and if they aren’t maximising their edges, the game will evolve and they’ll stand still.

Ayet: “Who’s your favorites for this year's world cup Dan and if you're allowed to say which players in your opinion should replace Rohit and Kohli or capable of starting at 1 or 3 instead of those players”

I don’t really have a favourite team, there are probably three teams who are stronger than others (Australia at home, plus England and India). Pakistan are also good. I’d be surprised if the winner didn’t come from those four teams.

My next article will be focused on a theme related to the Rohit and Kohli discussion so I’ll save my thoughts on that for the upcoming post.

Rashmi Patel: “Would like to know the codes and how it helps the players on field.”

Codes to players are becoming more utilised in cricket, although it’s still in the early stages. I’m sure most readers will have seen Nathan Leamon’s mystery letters/numbers which he showed at England matches, and that’s just one example.

I think it’s a good thing because players, and in particular, captains, need to retain an impossible amount of information in advance of a match, so helping them when they are on the pitch can be really useful. Let’s say you’re a bowler, and you’re trying to think of what the best ball is to the particular batter on strike - if someone is showing you the best option somehow then you can focus all your attention towards executing the plan.

The Secret Cricket Coach: “How do you develop good decision making around what the data shows? What I mean is everyone may have the same data but draw very different conclusions. The best will draw the most effective conclusions. How?”

This is a great question but also quite difficult to answer. There’s been some discussion about democratisation of data similar to some US sports (everyone can get the same data) and whether that creates a zero-sum game - I don’t think it does.

If you’re the world’s richest person and you decide to buy the world’s most expensive aeroplane, when it’s delivered to your house the first thing you are going to say is ‘how do I fly this thing?’. That’s where your pilot comes in, and the world’s richest person who owns the world’s most expensive aeroplane is also likely to want to recruit the world’s best pilot.

The analogy fits with cricket data. The best ‘pilots’ are the ones who can draw the most effective conclusions, and teams should, in theory, be fighting over them to 1) give them the biggest edges and 2) stopping rival teams from having those edges.

A transfer market for analysts isn’t far away…

* Thanks so much to everyone who has bought me a coffee. It is much appreciated! *